From Clinfowiki
Revision as of 03:37, 30 April 2017 by Erikstorm (Talk | contribs)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search


The role of the EHR with respect to imaging services has historically been to function as an interface to the Radiology Information System (RIS) for study ordering and retrieval of results. More mature implementations typically provide additional but limited functionality to view images via hyperlink within an external PACS (Picture Archiving and Communications System) viewer. While this system has functioned relatively well to date with respect to traditional DICOM radiology imaging services obtained within the context of the local enterprise, a wealth of important and information-rich multimedia clinical imaging present both within and extrinsic to the enterprise is not typically accessible through most current EHR implementations. Such (typically non-DICOM) clinical content is routinely acquired at point-of-care in both the inpatient and outpatient settings by multiple disciplines. Examples include photographic images of a patient's skin cancer taken during an outpatient dermatology visit (imaging informatics), ophthalmic imaging of the retina, digitally-captured images and/or video during endoscopic or orthopedic procedures, and forensic imaging taken during evaluation of suspected sexual or child abuse cases. A standardized workflow for the acquisition, storage, indexing, retrieval, and security of this nontraditional content is typically not homogeneous across an enterprise, therefore limiting its availability within most current EMR systems. A model for Enterprise Imaging is emerging as a method to optimize the secure capture and storage of all relevant clinical imaging across the enterprise, and to facilitate the sharing of this content via an integrated EHR viewer.

Sources of Imaging Content

The acquisition and primary interpretation of medical images was historically considered the purview of radiology, with conventional film radiography representing the bulk of imaging procedures. The spectrum of available imaging studies has now exploded to include both contrast and non-contrast computed tomography (CT), magnetic resonance imaging (MRI) and ultrasound (US), as well as the functional studies of nuclear medicine. Imaging is no longer isolated to the radiology department. Clinical providers from almost every discipline now frequently engage in some form of point-of-care imaging, including optical static or cine imaging obtained during endoscopic and arthroscopic procedures, and medical photography.

Workflow for Image Acquisition and Management

Multimedia clinical content is being acquired with increasing frequency and volume by numerous specialties across the medical enterprise, outside the traditional domains of radiology. While much of this imaging is currently obtained via the orders-based/SWF model using traditional modalities (e.g., the use of ultrasound by cardiology and OB/GYN), a growing volume of non-traditional “visible light” modality imaging is obtained by other specialties such as dermatology, ophthalmology, dentistry, and endoscopy. Imaging in these specialties is often obtained ad-hoc or incidentally during the clinical encounter within the context of a locally-defined workflow model, and is typically not the primary purpose for the visit (1). The purpose of imaging by these specialties could be to complement the clinical visit or procedure, or to document relevant findings for follow-up and/or billing purposes. Example use cases would include visible-light photography of a suspicious lesion being examined in a primary care or dermatology clinic, bedside ultrasound (FAST scan) performed in the ER for trauma, or digital capture images of a suspicious lesion observed during an endoscopic procedure. Similar to the transition in radiology from film-based to digital storage, the digitization of pathology will enable review of the relevant images along with the diagnostic report within the EHR. Concerns have been raised, however, regarding the lack of a DICOM standard (2). The ability to organize, index, store, secure, and retrieve this emerging content across the clinical enterprise requires adoption of a coherent standards-based model.

The typical workflow associated with this and similar types of multimedia content, assuming a defined workflow even exists, is typically heterogeneous across the clinical enterprise. Imaging of this nature, not originated through a traditional EHR order system, is considered unsolicited and typically does not require a separate, dedicated image interpretation report for billing purposes. Relevant image findings or descriptions thereof may instead be incorporated into the applicable progress or procedure notes within the EHR directly or through third-party applications. Acquired image content is often stored locally by the provider, either on an in-office computer hard drive or local area network storage device, and most often is not readily available or searchable across the broader enterprise via the EHR or other systems. Application of a more traditional, orders-based model to facilitate the integration of imaging content across the enterprise is not considered a logistically viable solution because of the intrinsically ad hoc nature of image acquisition in these settings (1). Solutions including the use of order sets automatically created along with a scheduled encounter are not typically feasible due to the fact that it may not be known prior to the encounter what imaging, if any, will be needed. The use of discrete and specific image orders of appropriate granularity of metadata to enable meaningful search and retrieval would be required in order to correctly index the image content within the EHR with respect to image type, anatomic location, etc. It would not be useful, for example, to have a photograph of a wound obtained to monitor healing if the patient has numerous other clinical photos indexed in their record simply as type: “Clinical Photograph”. More robust metadata descriptors could include “Clinical Photograph, Left Hand”, or “Digital Capture, Arthroscopic, Right Knee”, etc. A robust granularity would be to also include designation of anatomic position (e.g., anterior vs. lateral). The correct designation body part has been described as being likely the most important piece of metadata for enabling search and sorting across specialties, and a standard ontology for the purposes of body part mapping has been advocated (3). To achieve interoperability, this data along with appropriate patient identifiers must be communicated to, or entered directly at the modality, reconciled with other patient identifiers that may be used elsewhere within the enterprise (typically by assignment of a unique master patient index (MPI), and along with the acquired images be routed back to the EHR using an appropriate standards-based methods such as DICOM and HL7.

EHR Image Viewing Considerations

The goal of a robust and successfully-implemented enterprise imaging system is the ability to access relevant image and multimedia content consolidated in logical fashion within the electronic health record in context with other clinical documentation at the point-of-care. The concept of an integrated image viewer within the EHR is not new; departmental PACS applications frequently include the ability to view DICOM images via hyperlink or similar method within an enabled thin-client-type viewer application. This functionality was adequate for clinical review of standard-modality images in the context of a traditional order-based framework, but is unsuitable to serve the disparate need of clinicians seeking to view non-traditional and encounter-based multimedia content. This need can instead be met through inclusion of a multi-specialty “universal” enterprise viewer capable of displaying a broad range of content types. The collaborative SIIM-HIMSS member workgroup on Enterprise Imaging defines an enterprise viewer as “a thin-client or zero-client application used on any off-the-shelf device to distribute, display, and manipulate multi-specialty image, video, audio, and scanned documents stored in separate centralized archives through, or standalone from, the EHR” (4).

Characteristics of an Enterprise EHR Viewer

Viewing of imaging-related content within the EHR including associated diagnostic reports is performed as part of daily inpatient and ambulatory clinical activity by multiple stakeholders throughout the enterprise, including physician and nursing staff, physician extenders, technologists, and more recently patients. Each user has their own need for image consumption, be it for diagnostic interpretation or clinical review, and a successful integrated viewing platform will provide access to a broad range of tools, interface options, and speed to enable the appropriate degree of interaction required to support the given use case.

Given the limitations of EHR-integrated PACS thin-client, and other EHR viewing solutions, the alternative to a universal viewer may require support for multiple viewers specific to individual content types and associated integration profiles to allow interoperability with disparate departmental and/or specialty archives. The implementation of an enterprise viewer, along with a scalable, consolidated VNA, allows simplification of network architecture and a single system through which the EHR must search to find relevant content (5).

The EHR is accessed on multiple device platforms, including desktop computers, laptops, tablets, and even smartphones (Khanna, 2016). To facilitate speed of data transmission on mobile devices, modern enterprise viewers employ convert lossless DICOM objects into lossy or lossless non-DICOM formats (4). For the case of a file containing a large number of images, such as an echocardiogram or CT angiogram, speed of transmission is enhanced by conversion to a compressed/lossy format such as JPEG using discrete cosine transfer. This method would be appropriate for non-diagnostic image review. The alternative JPEG2000 format allows for either lossless (or reversible lossy) compression via an integer wavelet transform, and would be appropriate for diagnostic image review and/or streaming of large data sets (6).

EHR Viewer Security

An integrated EHR viewer will ideally launch within the context of a secure Uniform Resource Locator (URL) through an API (Application Programming Interface) at the imaging study level, thus becoming a seamless extension of the EHR. Access would ideally be made through existing EHR authentication, permitting application of access controls and permissions appropriate to the individual user's role. Security is enhanced if images are stripped of patient-identifying metadata, including conversion from DICOM to non-DICOM format if applicable, before being stored on the local device. Permissions may be required to access certain image content deemed to be “sensitive”, e.g., medical photography documentation of sexual or child abuse. Similar to other EHR activities, the use of the image viewer would be subject to logging and the generation of audit records using the [IHE] Audit Trail and Node Authentication (ATNA) profile to ensure appropriate unionization and HIPPA compliance (7).


1. Cram, D., Roth, C.J. & Towbin, A.J. J Digit Imaging (2016) 29: 559. doi:10.1007/s10278-016-9888-7

2. Kalinski, T., Zwönitzer, R., Roßner, M., Hofmann, H., Roessner, A., Guenther, T. (2012). Digital imaging and communications in medicine (DICOM) as standard in digital pathology. Histopathology, 61: 132–134. doi:10.1111/j.1365-2559.2012.04243.x

3. Towbin, A.J., Roth, C.J., Bronkalla, M. et al. J Digit Imaging (2016) 29: 574. doi:10.1007/s10278-016-9897-6

4. Roth, C.J., Lannum, L.M. & Joseph, C.L. J Digit Imaging (2016) 29: 539. doi:10.1007/s10278-016-9883-z

5. Oosterwijk, H. (2012). What Is a VNA, Anyway? [PDF]. Available at http://otechimg.com/publications/pdf/What_is_a_VNA_anyway.pdf.

6. Branstetter, B.F., Ed. (2009). Practical imaging informatics. New York, NY: Springer, pp 136. Digital Imaging and Communications in Medicine (DICOM). (2013).

7. Integrating the Healthcare Enterprise. (2015). Audit trail and node authentication. Retrieved from the IHE Wiki: http://wiki.ihe.net/index.php/audit_trail_and_node_authentication.

Additional Information

The complete series of white papers by the HIMSS-SIMM Enterprise Imaging workgroup is available from http://siim.org/page/himss_siim_white_pap.

Submitted by (Erik S. Storm, DO)